Objectives: 1. To do a natural history study of large bowel cancer in a number of insitutions cooperating in innovative treatment studies under the auspices of the National Large Bowel Cancer Project. 2. To establish a common data base of clinical characteristics of patients with colorectal cancer. 3. To develop statistical criteria in order to (a) aid in the assignment of treatment based on risk and (b) allow for rapid and accurate evaluation of new and innovative treatment approaches. This study will be a coordinated effort between each of the institutions and the Statistical Unit, Headquarters office of the Nation Large Bowel Cancer Project. A co-investigator, designated from each participating institution will be responsible for retrospective review of patient's records with colorectal cancer and coding of appropriate data onto forms for submission to the Statistical Unit. The Statistical Unit will be responsible for developing a central data base of this information and providing statistical analyses meant to define those characteristics which have prognostic significance. Various statistical techniques will be used such as multiple regression analysis, survival analysis, and discriminant function analysis. The identification and quantification, pretreatment of patient characteristics with independent prognostic significance will facilitate the identification of individual patients at varying risks by more precise sub-group classifications. This will allow for uniformity among the cooperating institutions both in the allocation of treatment and evaluation of results. By establishing historical controls, they can be matched to each patient group undergoing new treatment based on prognosis and will allow for more rapid evaluation of treatment effectiveness.